Main Findings
In a well-known pioneering study Eysenck used national insurance data (Equitable Life Society) to show that the outcome of ‘neurosis’, as defined at that time, was the same as that achieved by formal psychotherapy. In both groups, about third of the people in each group improved and the rest either remained the same or became worse [47]. The results of the Nottingham Study were very similar to those of Eysenck after 30 years. Close to a half still had a significant DSM diagnosis at assessment, with those with personality pathology having worse outcomes [23]. There was also little evidence that treatments given over the course of the study had major impact on outcome [46].
In interpreting the data showing about one third of patients have good outcomes, and why a minority have worse ones, the impact of environmental change features strongly. Those with positive environmental events had most of their pathology halved compared with those who had negative environmental events at 12 years and although the differences were less significant at 30 years account has to be taken of the smaller numbers in the 30 year follow up (Table 3).
The results give a strong indication that environmental factors are at least as effective, if not greater, than specific treatment interventions in common mental disorders. The clinical status of the patients overall was no better after 30 years than at baseline and yet there was clear evidence that different environmental changes across the range of major change behaviours had significant effects on outcome. Although it could be claimed that the substance use changes constituted treatment but in the cases where there was major positive change the patients regarded the advice about smoking and drug behaviour, usually given in the first or second assessment as critical in their improvement. It is also notable that four of the patients in the study specified the initial advice about smoking and drinking as their preferred treatment after 30 years. The fact that this particular memory was so important after this long period indicates the potential power of the first psychiatric interview, when a comprehensive assessment is followed by a formulation of the problem in all its aspects.
The smaller number of positive environmental changes in those with personality disorder could have contributed to the generally worse outcome in this group, a finding that is now very well known but not fully understood [48]. These patients also incurred greater costs than others at 12 years [49]. It is reasonable to suppose that if more positive events can be generated in those with a tendency to negative interactions, as in personality disorder, these consequences could be reversed.
Strengths And Limitations
Although there are many studies on the impact of the environment on health, with a growing number promoting access to nature and green spaces, these are focused mainly on general well-being rather than direct management of significant mental illness. They also focus on one type of environmental change instead of the full range available. Our study is unusual in three respects; (a) examining all types of environmental change, (b) having simultaneous records of changes in environment and clinical status, and (c) following the trajectories of patients for 30 years. There is one distant comparison. In a systematic review of health promotion in schools the effects of interventions focused on environmental change, nutrition, exercise, safety, substance use and family life were compared. Those promoting mental health rather than trying to prevent ill-health, and those focusing on environmental change were the most effective [50]. This accords with our findings but the studies are not strictly comparable.
There are several important limitations to this study. The systematic examination of types of environmental change, particularly the separation of instrumental and imposed change, was not part of the original study and necessarily much of the data have been obtained retrospectively. It could therefore be influenced by bias, but we have tried to keep this to a mimimum by allowing patients to drive the separation of positive from negative changes from their own reports and not by inference. By taking a longer time frame, interventions that might first be perceived as negative or neutral may be understood to be positive in the long-term, and vice-versa. This is the main advantage of a longer follow-up period.
Role of therapeutic intervention in positive instrumental change
Positive instrumental environmental change can be assisted by health practitioners in many ways. At its simplest level simple advice may be sufficient, such as explaining the advantages of giving up smoking or reducing hazardous drinking. At the other extreme someone with social disadvantage with very limited resources has very few options available and successful ones have to be worked for assiduously. Some of the interventions which led to environmental change in the Nottingham study were clearly therapist engendered. Although the benefits of these were too small to be significant they do suggest that if formal environmental interventions such as nidotherapy had been given during the course of the 30 years greater improvement might have been possible. Often the benefits of nidotherapy are delayed [51] and so it is preferable to use a long time course In their evaluation.
In this study there was no attempt to manipulate the environment deliberately to improve significant mental illness. It is becoming clear that this type of intervention could be of great value but at present has not been well studied. The two most common forms of environmental intervention are social prescribing and nidotherapy. In social prescribing patients can be referred to a link worker with knowledge of environmental options in the immediate locality. Such interventions may often be of value to the patient but research is this area has not yielded strong results because of barriers to implementation and concern over training needs [52–54] and one recent meta-analysis suggested it was both clinically ineffective and cost-ineffective [55]. Nidotherapy overlaps with social prescribing but is more complex [56], and has been shown to be of promise in several studies, including two randomised trials [51, 57–59] and also to improve social function in severe mental illness [60] but more larger scale studies are needed. The development of instrumental interventions more generally needs attention and will assist in this work.